What is in a voter's self-interest?

Alan Krueger reports on survey research that shows that people do not vote according to their self-interest. In particular, he bemoans the fact that a majority of the poor want to get rid of the estate tax. This and other odd results are due to "ignorance and uncertainty" says Larry Bartels, a Princeton political scientist. If only the poor were better informed they would vote against tax cuts for the rich. Moreover, a better informed electorate would be a good thing. I take issue with both of these positions.

Take the normative position first. Assuming that voters voted self-interestedly, would a more informed electorate be a good thing? Doubtful. If everyone voted their "interest," as Krueger and Bartels conceive it, every bureaucrat, welfare recipient and old person living on social security would vote for more government. Naturally, I think this would be a disaster but even those who think this would be a good thing ought to give pause when they consider how much more polarized our society would become were it not for the fact that ideology cuts across class lines.

Moreover, isn't it interesting that when the poor vote against their "self-interest" they are labeled "uninformed" - Bartels compares them to Homer Simpson. But when Hollywood liberals like Barbara Streisand or rich philanthropists like Bill Gates Sr. vote against their "self-interest" they are called enlightened. What Krueger and Bartels refer to as self-interest is actually masking an ideology.

Is it true that informed voters would vote differently? (Krueger cites some evidence suggesting that in fact this is not the case - at least not as much as one would expect - but he doesn't offer an explanation.) To understand this one should first realize that voters are uninformed because it doesn't pay to be informed. The probability that one vote sways the election is infinitesimal so voters are rationally ignorant. Does this imply that voting is random? Not at all. Voters who care about ideas even a little are free to vote their ideology at low cost. Thus, in my view, the fact that votes don't matter gives us hope. It's only because votes don't matter that libertarianism has a chance of success. Of course, I recognize that the same facts gives socialism a chance at the polls but I hope good ideas will win out.

Addendum: I've been influenced on these issues by our colleague, Bryan Caplan - although I give the ideas a more positive spin than he does. I recommend his paper Libertarianism Against Economism: How Economists Misunderstand Voters, and Why Libertarians Should Care from The Independent Review and his other papers on rational irrationality which you can find on his web page.